Ultrasonic Assessment of Liver Fibrosis Using One-Dimensional Convolutional Neural Networks Based on Frequency Spectra of Radiofrequency Signals with Deep Learning Segmentation of Liver Regions in B-Mode Images: A Feasibility Study.

The early detection of liver fibrosis is of significant importance. Deep learning analysis of ultrasound backscattered radiofrequency (RF) signals is emerging for tissue characterization as the RF signals carry abundant information related to tissue microstructures. However, the existing methods only used the time-domain information of the RF signals for liver fibrosis assessment, and the liver region of interest (ROI) is outlined manually. In this study, we proposed an approach for liver fibrosis assessment using deep learning models on ultrasound RF signals. The proposed method consisted of two-dimensional (2D) convolutional neural networks (CNNs) for automatic liver ROI segmentation from reconstructed B-mode ultrasound images and one-dimensional (1D) CNNs for liver fibrosis stage classification based on the frequency spectra (amplitude, phase, and power) of the segmented ROI signals. The Fourier transform was used to obtain the three kinds of frequency spectra. Two classical 2D CNNs were employed for liver ROI segmentation: U-Net and Attention U-Net. ROI spectrum signals were normalized and augmented using a sliding window technique. Ultrasound RF signals collected (with a 3-MHz transducer) from 613 participants (Group A) were included for liver ROI segmentation and those from 237 participants (Group B) for liver fibrosis stage classification, with a liver biopsy as the reference standard (Fibrosis stage: F0 = 27, F1 = 49, F2 = 51, F3 = 49, F4 = 61). In the test set of Group A, U-Net and Attention U-Net yielded Dice similarity coefficients of 95.05% and 94.68%, respectively. In the test set of Group B, the 1D CNN performed the best when using ROI phase spectrum signals to evaluate liver fibrosis stages ≥F1 (area under the receive operating characteristic curve, AUC: 0.957; accuracy: 89.19%; sensitivity: 85.17%; specificity: 93.75%), ≥F2 (AUC: 0.808; accuracy: 83.34%; sensitivity: 87.50%; specificity: 78.57%), and ≥F4 (AUC: 0.876; accuracy: 85.71%; sensitivity: 77.78%; specificity: 94.12%), and when using the power spectrum signals to evaluate ≥F3 (AUC: 0.729; accuracy: 77.14%; sensitivity: 77.27%; specificity: 76.92%). The experimental results demonstrated the feasibility of both the 2D and 1D CNNs in liver parenchyma detection and liver fibrosis characterization. The proposed methods have provided a new strategy for liver fibrosis assessment based on ultrasound RF signals, especially for early fibrosis detection. The findings of this study shed light on deep learning analysis of ultrasound RF signals in the frequency domain with automatic ROI segmentation.

Information Entropy and Its Applications.

Ultrasound is a first-line diagnostic tool for imaging many disease states. A number of statistical distributions have been proposed to describe ultrasound backscattering measured from tissues having different disease states. As an example, in this chapter we use nonalcoholic fatty liver disease (NAFLD), which is a critical health issue on a global scale, to demonstrate the capabilities of ultrasound to diagnose disease. Ultrasound interaction with the liver is typically characterized by scattering, which is quantified for the purpose of determining the degree of liver steatosis and fibrosis. Information entropy provides an insight into signal uncertainty. This concept allows for the analysis of backscattered statistics without considering the distribution of data or the statistical properties of ultrasound signals. In this chapter, we examined the background of NAFLD and the sources of scattering in the liver. The fundamentals of information entropy and an algorithmic scheme for ultrasound entropy imaging are then presented. Lastly, some examples of using ultrasound entropy imaging to grade hepatic steatosis and evaluate the risk of liver fibrosis in patients with significant hepatic steatosis are presented to illustrate future opportunities for clinical use.

Liver Fibrosis Assessment Using Radiomics of Ultrasound Homodyned-K imaging Based on the Artificial Neural Network Estimator.

The homodyned-K distribution is an important ultrasound backscatter envelope statistics model of physical meaning, and the parametric imaging of the model parameters has been explored for quantitative ultrasound tissue characterization. In this paper, we proposed a new method for liver fibrosis characterization by using radiomics of ultrasound backscatter homodyned-K imaging based on an improved artificial neural network (iANN) estimator. The iANN estimator was used to estimate the ultrasound homodyned-K distribution parameters k and α from the backscattered radiofrequency (RF) signals of clinical liver fibrosis (n = 237), collected with a 3-MHz convex array transducer. The RF data were divided into two groups: Group I corresponded to liver fibrosis with no hepatic steatosis (n = 94), and Group II corresponded to liver fibrosis with mild to severe hepatic steatosis (n = 143). The estimated homodyned-K parameter values were then used to construct k and α parametric images using the sliding window technique. Radiomics features of k and α parametric images were extracted, and feature selection was conducted. Logistic regression classification models based on the selected radiomics features were built for staging liver fibrosis. Experimental results showed that the proposed method is overall superior to the radiomics method of uncompressed envelope images when assessing liver fibrosis. Regardless of hepatic steatosis, the proposed method achieved the best performance in staging liver fibrosis ≥F1, ≥F4, and the area under the receiver operating characteristic curve was 0.88, 0.85 (Group I), and 0.85, 0.86 (Group II), respectively. Radiomics has improved the ability of ultrasound backscatter statistical parametric imaging to assess liver fibrosis, and is expected to become a new quantitative ultrasound method for liver fibrosis characterization.

Detection of pediatric hepatic steatosis through ultrasound backscattering analysis.

OBJECTIVES: Hepatic steatosis has become a considerable concern in the pediatric population. The objective of this study was to evaluate the feasibility of using ultrasound Nakagami imaging to produce a parametric image for analyzing the echo amplitude distribution to assess pediatric hepatic steatosis. METHODS: A total of 68 pediatric participants were enrolled in healthy control (n = 26) and study groups (n = 42). Raw data from ultrasound imaging were acquired for each participant analysis using AmCAD-US, a software approved by the US Food and Drug Administration for ultrasound Nakagami imaging. The Nakagami parameters were compared with the hepatic steatosis index (HSI) and the steatosis grade (G0: HSI < 30; G1: 30 ≤ HSI < 36; G2: 36 ≤ HSI < 41.6; G3: 41.6 ≤ HSI < 43; G4: HSI ≥ 43) using correlation analysis, one-way analysis of variance (ANOVA), and receiver operating characteristic (ROC) curve analysis. RESULTS: The Nakagami parameter increased from 0.53 ± 0.13 to 0.82 ± 0.05 with increasing severity of hepatic steatosis from G0 to G4 and were significantly different between the different grades of hepatic steatosis (p < .05). The areas under the ROC curves were 0.96, 0.92, 0.85, and 0.82 for diagnosing hepatic steatosis ≥ G1, ≥ G2, ≥ G3, and ≥ G4, respectively. CONCLUSIONS: The Nakagami parameter value quantifies changes in the echo amplitude distribution of ultrasound backscattered signals caused by fatty infiltration, providing a novel, noninvasive, and effective data analysis technique to detect pediatric hepatic steatosis. KEY POINTS: • Ultrasound Nakagami imaging enabled quantification of the echo amplitude distribution for tissue characterization. • The Nakagami parameter increased with the increasing severity of pediatric hepatic steatosis. • The Nakagami parameter demonstrated promising diagnostic performance in evaluating pediatric hepatic steatosis.

Fetal Ultrasound Image Segmentation for Automatic Head Circumference Biometry Using Deeply Supervised Attention-Gated V-Net.

Automatic computerized segmentation of fetal head from ultrasound images and head circumference (HC) biometric measurement is still challenging, due to the inherent characteristics of fetal ultrasound images at different semesters of pregnancy. In this paper, we proposed a new deep learning method for automatic fetal ultrasound image segmentation and HC biometry: deeply supervised attention-gated (DAG) V-Net, which incorporated the attention mechanism and deep supervision strategy into V-Net models. In addition, multi-scale loss function was introduced for deep supervision. The training set of the HC18 Challenge was expanded with data augmentation to train the DAG V-Net deep learning models. The trained models were used to automatically segment fetal head from two-dimensional ultrasound images, followed by morphological processing, edge detection, and ellipse fitting. The fitted ellipses were then used for HC biometric measurement. The proposed DAG V-Net method was evaluated on the testing set of HC18 (n = 355), in terms of four performance indices: Dice similarity coefficient (DSC), Hausdorff distance (HD), HC difference (DF), and HC absolute difference (ADF). Experimental results showed that DAG V-Net had a DSC of 97.93%, a DF of 0.09 ± 2.45 mm, an AD of 1.77 ± 1.69 mm, and an HD of 1.29 ± 0.79 mm. The proposed DAG V-Net method ranks fifth among the participants in the HC18 Challenge. By incorporating the attention mechanism and deep supervision, the proposed method yielded better segmentation performance than conventional U-Net and V-Net methods. Compared with published state-of-the-art methods, the proposed DAG V-Net had better or comparable segmentation performance. The proposed DAG V-Net may be used as a new method for fetal ultrasound image segmentation and HC biometry. The code of DAG V-Net will be made available publicly on https://github.com/xiaojinmao-code/ .

Ultrasound Backscatter Envelope Statistics Parametric Imaging for Liver Fibrosis Characterization: A Review.

Early detection and diagnosis of liver fibrosis is of critical importance. Currently the gold standard for diagnosing liver fibrosis is biopsy. However, liver biopsy is invasive and associated with sampling errors and can lead to complications such as bleeding. Therefore, developing noninvasive imaging techniques for assessing liver fibrosis is of clinical value. Ultrasound has become the first-line tool for the management of chronic liver diseases. However, the commonly used B-mode ultrasound is qualitative and can cause interobserver or intraobserver difference. Ultrasound backscatter envelope statistics parametric imaging is an important group of quantitative ultrasound techniques that have been applied to characterizing different kinds of tissue. However, a state-of-the-art review of ultrasound backscatter envelope statistics parametric imaging for liver fibrosis characterization has not been conducted. In this paper, we focused on the development of ultrasound backscatter envelope statistics parametric imaging techniques for assessing liver fibrosis from 1998 to September 2019. We classified these techniques into six categories: constant false alarm rate, fiber structure extraction technique, acoustic structure quantification, quantile-quantile probability plot, the multi-Rayleigh model, and the Nakagami model. We presented the theoretical background and algorithms for liver fibrosis assessment by ultrasound backscatter envelope statistics parametric imaging. Then, the specific applications of ultrasound backscatter envelope statistics parametric imaging techniques to liver fibrosis evaluation were reviewed and analyzed. Finally, the pros and cons of each technique were discussed, and the future development was suggested.

Clinical Evaluation of Duchenne Muscular Dystrophy Severity Using Ultrasound Small-Window Entropy Imaging.

Information entropy of ultrasound imaging recently receives much attention in the diagnosis of Duchenne muscular dystrophy (DMD). DMD is the most common muscular disorder; patients lose their ambulation in the later stages of the disease. Ultrasound imaging enables routine examinations and the follow-up of patients with DMD. Conventionally, the probability distribution of the received backscattered echo signals can be described using statistical models for ultrasound parametric imaging to characterize muscle tissue. Small-window entropy imaging is an efficient nonmodel-based approach to analyzing the backscattered statistical properties. This study explored the feasibility of using ultrasound small-window entropy imaging in evaluating the severity of DMD. A total of 85 participants were recruited. For each patient, ultrasound scans of the gastrocnemius were performed to acquire raw image data for B-mode and small-window entropy imaging, which were compared with clinical diagnoses of DMD by using the receiver operating characteristic curve. The results indicated that entropy imaging can visualize changes in the information uncertainty of ultrasound backscattered signals. The median with interquartile range (IQR) of the entropy value was 4.99 (IQR: 4.98-5.00) for the control group, 5.04 (IQR: 5.01-5.05) for stage 1 patients, 5.07 (IQR: 5.06-5.07) for stage 2 patients, and 5.07 (IQR: 5.06-5.07) for stage 3 patients. The diagnostic accuracies were 89.41%, 87.06%, and 72.94% for ≥stage 1, ≥stage 2, and ≥stage 3, respectively. Comparisons with previous studies revealed that the small-window entropy imaging technique exhibits higher diagnostic performance than conventional methods. Its further development is recommended for potential use in clinical evaluations and the follow-up of patients with DMD.

Clinical Value of Information Entropy Compared with Deep Learning for Ultrasound Grading of Hepatic Steatosis.

Entropy is a quantitative measure of signal uncertainty and has been widely applied to ultrasound tissue characterization. Ultrasound assessment of hepatic steatosis typically involves a backscattered statistical analysis of signals based on information entropy. Deep learning extracts features for classification without any physical assumptions or considerations in acoustics. In this study, we assessed clinical values of information entropy and deep learning in the grading of hepatic steatosis. A total of 205 participants underwent ultrasound examinations. The image raw data were used for Shannon entropy imaging and for training and testing by the pretrained VGG-16 model, which has been employed for medical data analysis. The entropy imaging and VGG-16 model predictions were compared with histological examinations. The diagnostic performances in grading hepatic steatosis were evaluated using receiver operating characteristic (ROC) curve analysis and the DeLong test. The areas under the ROC curves when using the VGG-16 model to grade mild, moderate, and severe hepatic steatosis were 0.71, 0.75, and 0.88, respectively; those for entropy imaging were 0.68, 0.85, and 0.9, respectively. Ultrasound entropy, which varies with fatty infiltration in the liver, outperformed VGG-16 in identifying participants with moderate or severe hepatic steatosis (p < 0.05). The results indicated that physics-based information entropy for backscattering statistics analysis can be recommended for ultrasound diagnosis of hepatic steatosis, providing not only improved performance in grading but also clinical interpretations of hepatic steatosis.

Monitoring Microwave Ablation Using Ultrasound Echo Decorrelation Imaging: An ex vivo Study.

In this study, a microwave-induced ablation zone (thermal lesion) monitoring method based on ultrasound echo decorrelation imaging was proposed. A total of 15 cases of ex vivo porcine liver microwave ablation (MWA) experiments were carried out. Ultrasound radiofrequency (RF) signals at different times during MWA were acquired using a commercial clinical ultrasound scanner with a 7.5-MHz linear-array transducer. Instantaneous and cumulative echo decorrelation images of two adjacent frames of RF data were calculated. Polynomial approximation images were obtained on the basis of the thresholded cumulative echo decorrelation images. Experimental results showed that the instantaneous echo decorrelation images outperformed conventional B-mode images in monitoring microwave-induced thermal lesions. Using gross pathology measurements as the reference standard, the estimation of thermal lesions using the polynomial approximation images yielded an average accuracy of 88.60%. We concluded that instantaneous ultrasound echo decorrelation imaging is capable of monitoring the change of thermal lesions during MWA, and cumulative ultrasound echo decorrelation imaging and polynomial approximation imaging are feasible for quantitatively depicting thermal lesions.

Cutoff Values of Acoustic Radiation Force Impulse Two-Location Measurements in Different Etiologies of Liver Fibrosis.

BACKGROUND: Acoustic radiation force impulse (ARFI) imaging is a popular modality to measure liver fibrosis. ARFI selects optimal locations for measurement under imaging guiding. However, there are concerns on study locations and observers bias. To decrease the variations, ARFI at two locations was measured with standardized protocol. This study attempted to establish its cutoff values according to Metavir fibrosis score in different etiologies. METHODS: A consecutive series of patients who received liver histology study were prospectively enrolled. All cases had hemogram, liver biochemistry, viral markers, and ARFI two-location measurements within 4 weeks of histology study. A standardized protocol was performed by single technologist. We excluded patients with alanine aminotransferase >5x upper limit normal. RESULTS: Five hundred and ten patients that included 153 seronegative for both HBsAg and anti-HCV Non-B non-C (NBNC), 33 autoimmune liver diseases (AILD), 261 chronic hepatitis B (CHB), and 63 chronic hepatitis C (CHC) were enrolled. About 83% of NBNC patients had fat cell >5%. For diagnosis of liver cirrhosis, the area under receiver operating characteristic curve of NBNC, AILD, CHB, and CHC groups was 0.937, 0.929, 0.784, and 0.937; the cutoff values for mean ARFI were 1.788, 2.095, 1.455, and 1.710 m/s, respectively. The sensitivity and specificity are both over 0.818 for patients with nonalcoholic fatty liver diseases, CHC, and AILD, but the corresponding data are only 0.727-0.756 in CHB. The Fibrosis-4 Score is as good as ARFI on fibrosis assessment in NBNC. CONCLUSION: The performance of ARFI two-location measurement is excellent in NBNC, AILD, and CHC, but is only satisfactory in CHB.

Ultrasound single-phase CBE imaging for monitoring radiofrequency ablation.

Radiofrequency (RF) ablation (RFA) is the most commonly used minimally invasive procedure for thermal ablation of liver tumors. Ultrasound not only provides real-time feedback of the electrode location for RFA guidance but also enables visualization of the tissue temperature. Changes in backscattered energy (CBE) have been widely applied to ultrasound temperature imaging for assessing thermal ablation. Pilot studies have revealed that significant shadowing features appear in CBE imaging and are caused by the electrode and RFA-induced gas bubbles. To resolve this problem, the current study proposed ultrasound single-phase CBE imaging based on positive CBE values. An in vitro model with tissue samples derived from the porcine tenderloin was used to validate the proposed method. During RFA with various electrode lengths, ultrasound scans of tissue samples were obtained using a clinical ultrasound scanner equipped with a convex array transducer of 3 MHz. Raw image data comprising 256 scan lines of backscattered RF signals were acquired for B-mode, conventional CBE, and single-phase CBE imaging by using the proposed algorithmic scheme. The ablation sizes estimated using CBE imaging and gross examinations were compared to calculate the correlation coefficient. The experimental results indicated that single-phase CBE imaging largely suppressed artificial CBE information in the shadowed region. Moreover, compared with conventional CBE imaging, single-phase CBE imaging provided a more accurate estimation of ablation sizes (the correlation coefficient was higher than 0.8).

Three-dimensional Visualization of Ultrasound Backscatter Statistics by Window-modulated Compounding Nakagami Imaging.

In this study, the window-modulated compounding (WMC) technique was integrated into three-dimensional (3D) ultrasound Nakagami imaging for improving the spatial visualization of backscatter statistics. A 3D WMC Nakagami image was produced by summing and averaging a number of 3D Nakagami images (number of frames denoted as N) formed using sliding cubes with varying side lengths ranging from 1 to N times the transducer pulse. To evaluate the performance of the proposed 3D WMC Nakagami imaging method, agar phantoms with scatterer concentrations ranging from 2 to 64 scatterers/mm3 were made, and six stages of fatty liver (zero, one, two, four, six, and eight weeks) were induced in rats by methionine-choline-deficient diets (three rats for each stage, total n = 18). A mechanical scanning system with a 5-MHz focused single-element transducer was used for ultrasound radiofrequency data acquisition. The experimental results showed that 3D WMC Nakagami imaging was able to characterize different scatterer concentrations. Backscatter statistics were visualized with various numbers of frames; N = 5 reduced the estimation error of 3D WMC Nakagami imaging in visualizing the backscatter statistics. Compared with conventional 3D Nakagami imaging, 3D WMC Nakagami imaging improved the image smoothness without significant image resolution degradation, and it can thus be used for describing different stages of fatty liver in rats.

Performance Evaluations on Using Entropy of Ultrasound Log-Compressed Envelope Images for Hepatic Steatosis Assessment: An In Vivo Animal Study.

Ultrasound B-mode imaging based on log-compressed envelope data has been widely applied to examine hepatic steatosis. Modeling raw backscattered signals returned from the liver parenchyma by using statistical distributions can provide additional information to assist in hepatic steatosis diagnosis. Since raw data are not always available in modern ultrasound systems, information entropy, which is a widely known nonmodel-based approach, may allow ultrasound backscattering analysis using B-scan for assessing hepatic steatosis. In this study, we explored the feasibility of using ultrasound entropy imaging constructed using log-compressed backscattered envelopes for assessing hepatic steatosis. Different stages of hepatic steatosis were induced in male Wistar rats fed with a methionine- and choline-deficient diet for 0 (i.e., normal control) and 1, 1.5, and 2 weeks (n = 48; 12 rats in each group). In vivo scanning of rat livers was performed using a commercial ultrasound machine (Model 3000, Terason, Burlington, MA, USA) equipped with a 7-MHz linear array transducer (Model 10L5, Terason) for ultrasound B-mode and entropy imaging based on uncompressed (HE image) and log-compressed envelopes (HB image), which were subsequently compared with histopathological examinations. Receiver operating characteristic (ROC) curve analysis and areas under the ROC curves (AUC) were used to assess diagnostic performance levels. The results showed that ultrasound entropy imaging can be used to assess hepatic steatosis. The AUCs obtained from HE imaging for diagnosing different steatosis stages were 0.93 (≥mild), 0.89 (≥moderate), and 0.89 (≥severe), respectively. HB imaging produced AUCs ranging from 0.74 (≥mild) to 0.84 (≥severe) as long as a higher number of bins was used to reconstruct the signal histogram for estimating entropy. The results indicated that entropy use enables ultrasound parametric imaging based on log-compressed envelope signals with great potential for diagnosing hepatic steatosis.

Ultrasound Entropy Imaging of Nonalcoholic Fatty Liver Disease: Association with Metabolic Syndrome.

Nonalcoholic fatty liver disease (NAFLD) is the leading cause of advanced liver diseases. Fat accumulation in the liver changes the hepatic microstructure and the corresponding statistics of ultrasound backscattered signals. Acoustic structure quantification (ASQ) is a typical model-based method for analyzing backscattered statistics. Shannon entropy, initially proposed in information theory, has been demonstrated as a more flexible solution for imaging and describing backscattered statistics without considering data distribution. NAFLD is a hepatic manifestation of metabolic syndrome (MetS). Therefore, we investigated the association between ultrasound entropy imaging of NAFLD and MetS for comparison with that obtained from ASQ. A total of 394 participants were recruited to undergo physical examinations and blood tests to diagnose MetS. Then, abdominal ultrasound screening of the liver was performed to calculate the ultrasonographic fatty liver indicator (US-FLI) as a measure of NAFLD severity. The ASQ analysis and ultrasound entropy parametric imaging were further constructed using the raw image data to calculate the focal disturbance (FD) ratio and entropy value, respectively. Tertiles were used to split the data of the FD ratio and entropy into three groups for statistical analysis. The correlation coefficient r, probability value p, and odds ratio (OR) were calculated. With an increase in the US-FLI, the entropy value increased (r = 0.713; p < 0.0001) and the FD ratio decreased (r = -0.630; p < 0.0001). In addition, the entropy value and FD ratio correlated with metabolic indices (p < 0.0001). After adjustment for confounding factors, entropy imaging (OR = 7.91, 95% confidence interval (CI): 0.96-65.18 for the second tertile; OR = 20.47, 95% CI: 2.48-168.67 for the third tertile; p = 0.0021) still provided a more significant link to the risk of MetS than did the FD ratio obtained from ASQ (OR = 0.55, 95% CI: 0.27-1.14 for the second tertile; OR = 0.42, 95% CI: 0.15-1.17 for the third tertile; p = 0.13). Thus, ultrasound entropy imaging can provide information on hepatic steatosis. In particular, ultrasound entropy imaging can describe the risk of MetS for individuals with NAFLD and is superior to the conventional ASQ technique.

A Review of Ultrasound Tissue Characterization with Mean Scatterer Spacing.

Tissues exhibiting quasi-periodic structures can be modeled as a collection of diffuse scatterers and coherent scatterers. The mean scatterer spacing (MSS) of coherent and quasi-periodic components is directly related to tissue microstructure and has become an important quantitative ultrasound (QUS) parameter in the characterization of quasi-periodic tissues. In this paper, a review of the literature on the development of MSS as a QUS parameter was conducted. First, a unified theoretical background of MSS estimates was provided. Then, the application of MSS estimates was summarized with respect to liver, spleen, breast, bone, muscle, and other tissues. MSS estimation techniques were applied to (a) the diagnosis of hepatitis, liver fibrosis and cirrhosis, and lesions in tissues such as liver, breast, and spleen; (b) the differentiation between benign and malignant breast tumors, and the grading of breast cancer; (c) the detection of cancellous bone; and (d) the monitoring of the efficacy of treatments such as thermal ablation, with various levels of success. Future developments were also discussed in terms of real-time implementation of MSS estimates, local MSS estimation, relationship of MSS to other QUS parameters, combination of MSS with other QUS parameters, in vivo validation of MSS estimates, MSS parametric imaging, and three-dimensional ultrasound tissue characterization.

A Computer-Aided Diagnosis Scheme For Detection Of Fatty Liver In Vivo Based On Ultrasound Kurtosis Imaging.

Fatty liver disease is a common disease caused by alcoholism, obesity, and diabetes, resulting in triglyceride accumulation in hepatocytes. Kurtosis coefficient, a measure of the peakedness of the probability distribution, has been applied to the analysis of backscattered statistics for characterizing fatty liver. This study proposed ultrasound kurtosis imaging as a computer-aided diagnosis (CAD) method to visually and quantitatively stage the fatty liver. A total of 107 patients were recruited to participate in the experiments. The livers were scanned using a clinical ultrasound scanner with a 3.5-MHz curved transducer to acquire the raw ultrasound backscattered signals for kurtosis imaging. The kurtosis image was constructed using the sliding window technique. Experimental results showed that kurtosis imaging has the ability to visualize and quantify the variation of backscattered statistics caused by fatty infiltration. The kurtosis coefficient corresponding to liver parenchyma decreased from 5.41 ± 0.89 to 3.68 ± 0.12 with increasing the score of fatty liver from 0 (normal) to 3 (severe), indicating that fatty liver reduces the degree of peakedness of backscattered statistics. The best performance of kurtosis imaging was found when discriminating between normal and fatty livers with scores ≥1: the area under the curve (AUC) is 0.92 at a cutoff value of 4.36 (diagnostic accuracy =86.9 %, sensitivity =86.7 %, specificity =87.0 %). The current findings suggest that kurtosis imaging may be useful in designing CAD tools to assist in physicians in early detection of fatty liver.

Artifact reduction of ultrasound Nakagami imaging by combining multifocus image reconstruction and the noise-assisted correlation algorithm.

Several studies have investigated Nakagami imaging to complement the B-scan in tissue characterization. The noise-induced artifact and the parameter ambiguity effect can affect performance of Nakagami imaging in the detection of variations in scatterer concentration. This study combined multifocus image reconstruction and the noise-assisted correlation algorithm (NCA) into the algorithm of Nakagami imaging to suppress the artifacts. A single-element imaging system equipped with a 5 MHz transducer was used to perform the brightness/depth (B/D) scanning of agar phantoms with scatterer concentrations ranging from 2 to 32 scatterers/mm(3). Experiments were also carried out on a mass with some strong point reflectors in a breast phantom using a commercial scanner with a 7.5 MHz linear array transducer operated at multifocus mode. The multifocus radiofrequency (RF) signals after the NCA process were used for Nakagami imaging. In the experiments on agar phantoms, an increasing scatterer concentration from 2 to 32 scatterers/mm(3) led to backscattered statistics ranging from pre-Rayleigh to Rayleigh distributions, corresponding to the increase in the Nakagami parameter measured in the focal zone from 0.1 to 0.8. However, the artifacts in the far field resulted in the Nakagami parameters of various scatterer concentrations to be close to 1 (Rayleigh distribution), making Nakagami imaging difficult to characterize scatterers. In the same scatterer concentration range, multifocus Nakagami imaging with the NCA simultaneously suppressed two types of artifacts, making the Nakagami parameter increase from 0.1 to 0.8 in the focal zone and from 0.18 to 0.7 in the far field, respectively. In the breast phantom experiments, the backscattered statistics of the mass corresponded to a high degree of pre-Rayleigh distribution. The Nakagami parameter of the mass before and after artifact reduction was 0.7 and 0.37, respectively. The results demonstrated that the proposed method for artifact reduction allows a sensitive and effective scatterer characterization by Nakagami imaging.

Classification of Benign and Malignant Breast Tumors in Ultrasound Images with Posterior Acoustic Shadowing Using Half-Contour Features.

Posterior acoustic shadowing (PAS) can bias breast tumor segmentation and classification in ultrasound images. In this paper, half-contour features are proposed to classify benign and malignant breast tumors with PAS, considering the fact that the upper half of the tumor contour is less affected by PAS. Adaptive thresholding and disk expansion are employed to detect tumor contours. Based on the detected full contour, the upper half contour is extracted. For breast tumor classification, six quantitative feature parameters are analyzed for both full contours and half contours, including standard deviation of degree (SDD), which is proposed to describe tumor irregularity. Fifty clinical cases (40 with PAS and 10 without PAS) were used. Tumor circularity (TC) and SDD were both effective full- and half-contour parameters in classifying images without PAS. Half-contour TC [74 % accuracy, 72 % sensitivity, 76 % specificity, 0.78 area under the receiver operating characteristic curve (AUC), p > 0.05] significantly improved the classification of breast tumors with PAS compared to that with full-contour TC (54 % accuracy, 56 % sensitivity, 52 % specificity, 0.52 AUC, p > 0.05). Half-contour SDD (72 % accuracy, 76 % sensitivity, 68 % specificity, 0.81 AUC, p < 0.05) improved the classification of breast tumors with PAS compared to that with full-contour SDD (62 % accuracy, 80 % sensitivity, 44 % specificity, 0.61 AUC, p > 0.05). The proposed half-contour TC and SDD may be useful in classifying benign and malignant breast tumors in ultrasound images affected by PAS.

Semi-automatic breast ultrasound image segmentation based on mean shift and graph cuts.

Computerized tumor segmentation on breast ultrasound (BUS) images remains a challenging task. In this paper, we proposed a new method for semi-automatic tumor segmentation on BUS images using Gaussian filtering, histogram equalization, mean shift, and graph cuts. The only interaction required was to select two diagonal points to determine a region of interest (ROI) on an input image. The ROI image was shrunken by a factor of 2 using bicubic interpolation to reduce computation time. The shrunken image was smoothed by a Gaussian filter and then contrast-enhanced by histogram equalization. Next, the enhanced image was filtered by pyramid mean shift to improve homogeneity. The object and background seeds for graph cuts were automatically generated on the filtered image. Using these seeds, the filtered image was then segmented by graph cuts into a binary image containing the object and background. Finally, the binary image was expanded by a factor of 2 using bicubic interpolation, and the expanded image was processed by morphological opening and closing to refine the tumor contour. The method was implemented with OpenCV 2.4.3 and Visual Studio 2010 and tested for 38 BUS images with benign tumors and 31 BUS images with malignant tumors from different ultrasound scanners. Experimental results showed that our method had a true positive rate (TP) of 91.7%, a false positive (FP) rate of 11.9%, and a similarity (SI) rate of 85.6%. The mean run time on Intel Core 2.66 GHz CPU and 4 GB RAM was 0.49 ± 0.36 s. The experimental results indicate that the proposed method may be useful in BUS image segmentation.

Changes in mechanical properties at the muscle level could be detected by Nakagami imaging during in-vivo fixed-end contractions.

In this study, we investigated the capability of the Nakagami transformation to detect changes in vastus lateralis muscle-tendon stiffness (k) during dynamic (and intense) contractions. k was evaluated in eleven healthy males using the gold-standard method (a combination of ultrasound and dynamometric measurements) during maximal and sub-maximal voluntary fixed-end contractions of the knee extensors (20, 40, 60, 80, and 100% of maximum voluntary force), while Nakagami parameters were analysed using the Nakagami transformation during the same contractions. Muscle-belly behaviour was investigated by means of B-mode ultrasound analysis, while Nakagami parameters were obtained in post-processing using radiofrequency data. k was calculated as the slope of the force-muscle-belly elongation relationship. Three contractions at each intensity were performed to calculate the intra-trial reliability and the coefficient of variation (CV) of the Nakagami parameters. At all contraction intensities, high values of intra-trial reliability (range: 0.92-0.96) and low CV (<9%) were observed. k and Nakagami parameters increased as a function of contraction intensity, and significant positive correlations were observed between these variables. These data suggest that changes in mechanical properties (e.g., stiffness) at the muscle level could be investigated by means of Nakagami parameters.